Willingness to Use Instant Home HIV Tests Data from the California Behavioral Risk Factor Surveillance Survey

Transcription

1 Willingness to Use Instant Home HIV Tests Data from the California Behavioral Risk Factor Surveillance Survey Kathryn A. Phillips, PhD, James L. Chen, MPH Objective: Methods: Results: Conclusion: Although home collection HIV tests are available in the United States, home HIV tests providing instant results are not available. The objective of this study was to examine willingness to use instant home HIV tests and what test characteristics are most important to individuals. Six questions were added to the 1999 California Behavioral Risk Factor Surveillance Survey (BRFSS) (n 2964). This state-representative survey is part of the ongoing, 50-state BRFSS sponsored by the Centers for Disease Control and Prevention. Chi-square tests and logistic regression were used to examine willingness to use instant home HIV tests and preferences for HIV test characteristics. Over one third (37%) of respondents would consider using an instant home HIV test. Respondents willing to use instant home tests were more likely to be Hispanic (odds ratio [OR] 1.55; confidence interval [CI] ); to have not previously been tested for HIV (OR 1.72; CI ); to be more likely to plan to be tested in the next 12 months (OR 1.17; CI ); and to prefer urine sample collection (OR 1.56; CI ). Over one third of respondents in a large, state-representative sample would consider using an instant home HIV test. Hispanics, individuals who planned to be tested in the next 12 months, and individuals who had never been tested were more willing to use instant home tests. (Am J Prev Med 2003;24(4): ) 2003 American Journal of Preventive Medicine Introduction Despite decades of HIV prevention efforts, increasing the numbers of people knowing their HIV status continues to be a major public health challenge in the United States. 1 Nationally, almost one third of those infected with HIV do not know their HIV status. 2,3 Individuals who are at risk and do not test for HIV represent lost opportunities for HIV prevention through HIV counseling, risk reduction, and referral services. Furthermore, many HIV-positive individuals are diagnosed late in their illness and do not receive the medical benefits of early HIV therapy. 4 There are many potential barriers to HIV testing, including concerns about anonymity and privacy, the need for a blood draw by trained personnel, and the typical 2-week wait for test results. Thus, new HIV From the University of California, San Francisco (Phillips), San Francisco, California; and University of California, Davis (Chen), Davis, California JLC was affiliated with the University of California, San Francisco when the majority of this work was conducted. Address correspondence and reprint requests to: Kathryn A. Phillips, PhD, School of Pharmacy and Institute of Health Policy Studies, University of California, San Francisco, 3333 California St., Suite 420, Box 0613, San Francisco CA kathryn@itsa.ucsf.edu. testing technologies have been developed to increase testing utilization and overcome barriers to testing One new testing technology designed to overcome some of these barriers is the home collection HIV test, which was approved by the U.S. Food and Drug Administration (FDA) in Home collection tests allow persons to test themselves for HIV at home without medical supervision and to receive results over the telephone. However, the currently available home collection test in the United States requires a 3- to 7-day wait for results and a fingerstick sample, which may be less preferred by some individuals than other noninvasive methods. Other new tests use additional noninvasive sample collection methods that do not require trained personnel to administer, including an oral mucosal transudate swabbing test (oral fluids test) and a urine sample test. Lastly, rapid tests have been developed to decrease the waiting time for results to 1 hour. Rapid tests address the problem that many people tested do not return for their results. A national study of HIV testing at publicly funded sites revealed that 25% of persons testing positive and 33% of those testing negative did not return for their test results. 16 However, currently available rapid tests in the United States cannot be used at 340 Am J Prev Med 2003;24(4) /03/$ see front matter 2003 American Journal of Preventive Medicine Published by Elsevier Inc. doi: /s (03)

2 home because they require venipuncture and trained laboratory personnel. The FDA has been considering the approval of several new rapid HIV tests. These tests, including tests based on oral fluids and blood, are designed to be accurate and simple, to have a limited number of steps, and to be performed without trained laboratory staff. Many of these tests are currently being used in other countries, particularly developing countries with limited laboratory facilities. In November 2002, the FDA approved the first rapid, point-of-care HIV-1 test, OraQuick. 17 The rapid tests currently under consideration in the United States are for clinic-based use only and not being proposed for home use. However, the development of these tests signals that instant home tests are a possibility and could be proposed in the near future. 18 As noted by one observer: The new rapid tests will eventually find their way not only into the U.S. market, but into home-test kits as well. It s simple: Consumers want to know. 19 This study fills a gap in what is known about preferences for new types of HIV tests. There has been virtually no research on preferences for instant home HIV tests. Preferences for instant home tests were previously examined in a survey of individuals being tested at public clinics in San Francisco, California. Nearly one quarter (24%) of respondents indicated that they would choose an instant home test as their most preferred choice 20 and that instant home tests would become the most commonly preferred choice if they were highly accurate and cost only $ The objectives of this study were to examine (1) willingness to use instant home HIV tests, and (2) what HIV test characteristics are preferred by individuals. Data came from the Behavioral Risk Factor Surveillance Survey (BRFSS), an ongoing, well-established, validated survey used in all 50 states under direction of the National Center for Chronic Disease Prevention and Health Promotion of the Centers for Disease Control and Prevention. Data were representative of California, the most populous state of the United States at approximately 33.8 million people. 22 Methods Research Questions This study examined two research questions: 1. What are the characteristics of respondents who are willing to use instant home tests? It was expected that sociodemographics, prior HIV testing, plans to be HIV tested in the future, and HIV risk factors would be associated with preferences for instant home tests. In a previous study of preferences for HIV tests, it was found that whites, males, and higher-income people were more likely to prefer instant home tests. 20 A study of home collection tests found that users of home collection tests were more likely than public clinic testers to have not previously been tested. 14 Similarly, a study of willingness to use home collection tests prior to their availability found that males, individuals at higher risk, and those who had never been tested were more likely to be willing to use these tests; however, the study found that individuals with lower education and incomes were more likely to be willing to use home collection tests. 13 In this study, all respondents who might consider being tested were included, even if they did not perceive that they were at risk for HIV or did not plan to be tested in the next year. Including all respondents was important because we were interested in estimating future demand, and the instant home test might encourage individuals to be tested who would otherwise not do so. 2. Preferred test characteristics. Where do people want to be tested? What method of testing do people prefer? How quickly do they want results? Respondents were expected to prefer a doctor s office because it is the most common location for HIV testing, 16 but it was also expected that many would prefer a public clinic or to test at home. In addition, it was expected that respondents would prefer the most noninvasive methods (oral fluids or urine). Although a rapid result was expected to be preferred over waiting 2 weeks, the rapid test was described as requiring re-testing (which is the current standard); therefore, this characteristic was expected to decrease preference for rapid results. Data Source These data were obtained from the 1999 California (CA) BRFSS, a phone survey administered by the California Department of Health Services through random-digit dialing of California residences (response rate 82%). The resulting sample is representative of the California population. The survey was part of the national BRFSS developed by the Centers for Disease Control and Prevention (CDC) and conducted in each state every year since Each state s core survey remains consistent with the CDC questionnaire to obtain results that can be extrapolated to the general U.S. population. However, each state may also tailor questions to fit its specific population. This study used data from six unique questions that were added to the 1999 CA BRFSS questionnaire regarding HIV testing preferences. These data were then analyzed in conjunction with data from the standard BRFSS survey. The questions were added in July 1, 1999, for a 6-month period; therefore, data were obtained from 2964 respondents out of the total 4149 respondents in CA for The questions were asked of all participants aged 18 to 99 years. For descriptive analyses, the total sample was included. However, for regression analyses, the sample was restricted to ages 18 to 44 years because the questions in the core BRFSS survey on HIV testing were asked only of this age group (unweighted n 871). The six questions added to the survey were designed to efficiently answer research questions regarding preferences for HIV home testing kits (see Appendix). The first question asked about plans to be tested for HIV in the next 12 months, using wording from the National Health Interview Survey. 23 A 12-month time frame was used to obtain stronger measures of intentions. Respondents Am J Prev Med 2003;24(4) 341

3 Table 1. Sample characteristics %in sample (N 2964) Table 1. (continued) %in sample (N 2964) Sociodemographic characteristics and HIV testing history (BRFSS core survey) Gender Male 49 Female 51 Age (years) Education High school graduate 40 Technical school/some college 30 College graduate 30 Household income (annual) $20, $20,000 $49, $50, Race/ethnicity White 61 African American 7 Hispanic 22 Asian/Pacific Islander 8 American Indian 1 What are your personal chances of getting HIV? High 6 Medium 6 Low 29 None 57 Ever been tested for HIV? Yes 56 No 42 Have been tested for HIV in last 12 months? Yes 34 No 6 Where did you have last HIV test? Public clinic or setting 53 Private doctor s office 39 Home 2 HIV testing preferences (added to BRFSS) Do you think will be tested for HIV in next 12 months? Definitely yes 20 Probably yes 9 Maybe 10 Probably not 22 Definitely not 38 Where would you prefer to be tested? At public clinic 18 At doctor s office 70 At home using kit 9 None of these 1 How do you prefer being tested? Finger prick 60 Mouth swab 15 Urine 20 None of these 1 (continued) How soon would you want results? Within 2 weeks 59 At once, even if must repeat 39 Would you use an instant home test? Definitely yes 6 Probably yes 14 Maybe 17 Probably not 29 Definitely not 33 BRFSS, Behavioral Risk Factor Surveillance Survey. who would definitely not be tested were not asked further questions. The next three questions assessed preferences for location, method of sample collection, and speed of obtaining results. The location question included the three primary testing locations: public clinic, doctor s office, or at home. (Note that home location is considered as a separate attribute, as distinguished from a home test kit.) The method of collection question included three choices based on those that are currently feasible for home tests: blood sample by pricking the finger, oral fluid sample (described as putting a cotton swab in your mouth ), and a urine sample. The question on rapidity of results described instant results as possibly requiring a repeat test in order to ensure that the results are correct. The last two questions asked about instant home tests. Respondents were asked if they would be willing to use an instant home test if they were tested in the next 12 months. The test was described as providing results at once to distinguish instant home testing kits from home collection kits, and the question asked about intentions to test in the next 12 months in order to obtain a stronger measure of intentions. Respondents who answered definitely yes, probably yes, or maybe were asked the sixth question, which asked respondents their willingness to use an instant home test if it cost $50. Fifty dollars was chosen as the estimated maximum price, based on currently available home collection tests and discussion with testing experts and manufacturers. The six questions were reviewed and approved by the California Department of Health Services. Due to survey limitations, the number of questions and their length were limited. Statistical Analyses Chi-square tests were used to examine bivariate relationships, and logistic regression was used to examine the predictors of willingness to use a HIV home test kit. All analyses used sampling weights to reflect the CA population and standard error adjustment to account for the complex survey design (using Stata version 7.0 [Stata Corporation, College Station, Texas]). Hosmer Lemeshow goodness-of-fit tests were used to ensure that the logistic regression models adequately fit the data American Journal of Preventive Medicine, Volume 24, Number 4

4 Results Sample Characteristics The majority of respondents were aged 18 to 44 and non-hispanic white or Hispanic (Table 1). Forty percent had a high school education or less. Over 80% perceived themselves at low or no risk of contracting HIV, although 56% stated that they had tested before. The majority of respondents who had been previously tested did so at public clinics. About one third (39%) of respondents definitely, probably, or maybe planned to be tested for HIV in the next 12 months. The majority would prefer being tested at a doctor s office (70%), using the fingerstick method (60%), and waiting 2 weeks for results (59%). Over one third (37%) of respondents would definitely (6%), probably (14%), or maybe (17%) use an instant home test. Characteristics of Individuals Willing to Use Instant Home HIV Tests Respondents willing to use an instant home test were significantly more likely to have lower incomes, to be Hispanic, or to have not previously been tested (p 0.001, Table 2). They were also significantly more likely to choose a home location, a mouth swab or urine test (versus a fingerstick test), or rapid results (p 0.01). Surprisingly, there were no significant differences by perceived HIV risk or plans to be tested in the next 12 months, although there was a weak trend for respondents willing to use instant home tests if they were at higher perceived risk or planning to be tested (p 0.21 and p 0.45, respectively). Among respondents willing to use a home test, 24% would definitely or probably still be willing if the test were $50. In regression analyses, respondents willing to use instant home tests were more likely to be Hispanic (OR 1.55, CI ), to have never been tested for HIV (OR 1.72, ), to be more likely to plan to be tested in the next 12 months (OR 1.17, CI ), and to prefer urine sample collection (OR 1.56, CI ) (Table 3). As expected, respondents willing to use instant home tests were also more likely to prefer rapid results even if they may require retesting (OR 1.53, CI ) and to prefer a home location for testing (OR 9.70, CI ). Results were similar in sensitivity analyses using different forms of the model (available on request). Preferences for Test Characteristics There were several significant differences in preferences for test location (Table 4). For example, respondents who chose a home location were significantly more likely than respondents choosing a public clinic or doctor s office to prefer a mouth swab test, to have higher education, or to be white or African American. Although respondents preferring a home location had lower self-reported HIV risk than respondents preferring public clinics, they had higher levels of selfreported risk than respondents preferring doctors offices. There were also several significant differences by willingness to use different sample collection methods (Table 5). For example, respondents preferring a fingerstick test were more likely to prefer public clinics, to be willing to wait 2 weeks for results, and to be less likely to use an instant home test than respondents preferring a mouth swab or urine test. Respondents preferring a mouth swab test were more likely to prefer testing at home, to be female, to be more highly educated, and to be white than respondents preferring fingerstick or urine tests. Hispanics were more likely to prefer a fingerstick or urine test, while African Americans were more likely to prefer a urine test. Lastly, there were also statistically significant differences in preferences for when to obtain results (Table 6). Respondents preferring rapid tests were more likely to prefer a doctor s office or home location, a mouth swab or urine test, and to have either low or high income. Discussion This study is the first, to our knowledge, to use population-based data to examine the willingness of individuals to use instant home HIV tests. It was found that over one third of Californians would be willing to use instant home HIV tests, which is similar to the percentage of individuals tested for HIV in the past 12 months. Although it is unlikely that all of these individuals would actually use instant home HIV tests, and many would continue to use traditional testing methods and sites, these results suggest that instant home tests could meet the needs of substantial numbers of individuals. The results are also consistent with prior findings on the high interest of many individuals in new methods of HIV testing ,20,21 These findings have several implications for HIV testing practice and policy. The results for Hispanics and individuals who have never been tested are particularly striking. The odds for willingness to use instant home tests for Hispanics were 55% higher than for other racial/ethnic groups, and the odds for persons never tested were 72% higher than for persons previously tested. The preference for instant home tests by individuals who have never been tested for HIV indicates an opportunity to improve HIV testing rates among a population that does not utilize HIV testing services. The finding that Hispanics prefer instant home HIV tests reveals another population that could benefit from the development of instant HIV tests. 25 The high interest in instant home tests suggests that the development of such tests should become a priority for HIV prevention policy. As previously discussed, Am J Prev Med 2003;24(4) 343

5 Table 2. Characteristics of respondents willing to use instant home HIV tests Willing to use instant home test %(n) a (n 480) Characteristics and HIV testing history (BRFSS core survey) Gender Male 58 (226) Female 43 (254) Age (years) (67) (135) (145) (81) (33) (19) Education High school graduate 45 (205) Technical school/some college 28 (130) College graduate 27 (144) Household income (annual) $20, (169) $20,000 $49, (156) $50, (135) Race/ethnicity White 51 (241) African American 8 (35) Hispanic 32 (165) Asian/Pacific Islander 10 (36) What are your personal chances of getting HIV? High 10 (36) Medium 8 (29) Low 35 (120) None 47 (161) Ever been tested for HIV? Yes 51 (189) No 49 (157) Have been tested for HIV in last 12 months? p b Yes 32 (60) No 68 (129) Where did you have last HIV test? Public clinic or setting 55 (32) Private doctor s office 45 (25) Home 0 (0) Do you think will be tested for HIV in next 12 months? Definitely yes 15 (72) Probably yes 22 (102) Maybe 23 (110) Probably not 40 (192) Definitely not 1 (2) Where would you prefer to be tested? At public clinic 21 (97) At doctor s office 58 (269) At home using kit 21 (101) How do you prefer being tested? Finger prick 45 (195) Mouth swab 25 (132) Urine 30 (130) (continued) Table 2. (continued) instant home tests are technically feasible and are being used in other countries. However, the approval of such tests in the United States undoubtedly faces numerous barriers, as did home collection HIV tests 13 and is currently facing rapid tests. 26 Furthermore, although consumer preferences should be considered, these have to be balanced against public health needs; thus, the debate over the pros and cons of home tests should be continued. 17,26 Further research should be conducted to examine preferences for instant home tests and how much people are willing to pay for such tests, Table 3. Predictors of willingness to use instant home HIV tests (logistic regression, n 870, among respondents aged years) a Explanatory variables Willing to use instant home test %(n) a (n 480) How soon would you want results? Within 2 weeks 52 (244) At once, even if must repeat 48 (235) Would you buy an instant home test if it cost $50? c Definitely yes 7 (29) Probably yes 17 (72) Maybe 19 (96) Probably not 28 (133) Definitely not 31 (148) a Answered definitely, probably, or maybe. b Comparing respondents who definitely, probably, or may be willing to use instant home test versus probably or definitely not willing. c Number of respondents who would probably or definitely not use an instant home test were skipped for this question. BRFSS, Behavioral Risk Factor Surveillance Survey. OR (95% CI) Age (continuous, age 18 44) 1.01 ( ) Male (vs female) 1.18 ( ) Low income, $20,000 (vs other 1.10 ( ) income) Race/ethnicity African American (vs other race/ 0.64 ( ) ethnicity) Hispanic (vs other race/ethnicity) 1.55 ( )* Never tested before (vs tested before) 1.72 ( )** Plan to be tested in next 12 months 1.17 ( )* (ordinal, low to high intention) Prefer to test at home (vs elsewhere) 9.70 ( )*** Sample collection method Prefer mouth swab (vs fingerstick) 1.55 ( ) Prefer urine (vs fingerstick) 1.56 ( )* Prefer rapid results (vs 2 weeks) 1.53 ( )* *p 0.05; **p 0.01; ***p a Hosmer Lemeshow goodness-of-fit test, p 0.45; dependent variable 1ifdefinitely, probably, or maybe willing to use instant home HIV test. CI, confidence interval; OR, odds ratio. p b 344 American Journal of Preventive Medicine, Volume 24, Number 4

6 Table 4. Characteristics of respondents by preference for testing at doctor s office, clinic, or home location % preferring to be tested at: Public clinic Doctor office Home p Prefer to be tested how n 207 n 804 n Fingerstick Mouth swab Urine Prefer to know results when n 222 n 844 n Within 2 weeks At once, even if must repeat Would use instant home test kit n 222 n 855 n Definitely yes Probably yes Maybe Probably not Definitely not Age (years) n 223 n 861 n Education n 223 n 860 n High school graduate Some tech/college College graduate Household income (annual) n 208 n 827 n $20, $20,000 $49, $50, Race n 222 n 849 n White African American Hispanic Asian/Pacific Islander Personal chances of getting HIV n 173 n 595 n High Medium Low None Note: Only statistically significant variables (p 0.05) are shown. as well as the regulatory, industry, and social issues related to their approval. These results on preferences for different test characteristics are useful not only for developing new testing methods but also for improving current testing options. Although many individuals prefer being tested at doctors offices, many providers do not regularly offer HIV testing services. 27 Thus, continued emphasis on increasing the willingness of providers to offer HIV testing is critical. Although the majority of individuals preferred the use of fingersticks to obtain testing samples, over one third of respondents preferred mouth swab or urine tests, which are more noninvasive. We suspect that most individuals would prefer these noninvasive methods if they were more familiar with the methods and perceived them to be as accurate as blood tests. Overall, most respondents preferred to receive HIV test results in 2 weeks. This result is surprising given that the emphasis on development of rapid HIV tests assumes that individuals will want rapid results. However, the question was worded to indicate that retesting may be required because the currently available rapid test requires retesting to confirm initially positive results. In a previous study, 21 it was found that respondents have a strong dislike for rapid results that are less accurate and thus require retesting. Therefore, rapid tests are not likely to gain wide acceptance unless they are as accurate as current testing algorithms. This study had several limitations. The data are representative only of the CA population, which is more diverse in terms of race/ethnicity, income, and education than the overall United States. However, the population of CA may be more relevant for HIV testing issues because it possesses a larger percentage of individuals at risk for HIV than many other states. This study is also strengthened by being based on data from an ongoing validated survey. However, the BRFSS does Am J Prev Med 2003;24(4) 345

7 Table 5. Characteristics of respondents by willingness to use fingerstick, mouth swab, or urine tests % preferring to be tested using: Finger stick Mouth swab Urine p Prefer to be tested where: n 562 n 258 n Public clinic Doctor s office At home Prefer to know results when: n 679 n 264 n Within 2 weeks At once, even if must repeat Would use instant home test: n 570 n 264 n Definitely yes Probably yes Maybe Probably not Definitely not Gender n 958 n 264 n Male Female Education n 958 n 264 n High school graduate Some tech/college College graduate Race/ethnicity n 948 n 258 n White African American Hispanic Asian/Pacific Islander Note: Only statistically significant variables (p 0.05) are shown. Table 6. Characteristics of respondents by preferences for when to obtain test results % preferring to know results: At once, even if must be retested Within 2 weeks Prefer to be tested where: n 517 n Public clinic Doctor s office At home Prefer to be tested how: n 493 n Fingerstick Mouth swab Urine Household income (annual) n 509 n $20, $20,000 $49, $50, Only statistically significant variables (p 0.05) are shown. p not include detailed questions about HIV risk, and we were limited to adding brief questions about HIV testing preferences. However, several checks for comprehension suggested that respondents did understand the questions (details on request). These results are congruent with a previous survey and focus groups that obtained more detailed information on test preferences but in a narrower population. 20,28 Further research utilizing the BRFSS in other states could better gauge whether these results are generalizable to other areas of the United States. In conclusion, these results indicate that more than a third of Californians would be willing to use an instant home HIV test and that individuals who preferred instant tests were more likely to have never been tested for HIV. As the HIV epidemic continues, innovative testing technologies such as instant home HIV tests that provide reliable and accurate results at a reasonable cost should become a priority for HIV prevention strategies. Such technologies that simplify the HIV testing process will be needed to reach various populations that underutilize traditional testing venues. We gratefully acknowledge Dawn Flick, MPH, University of California-Berkeley, for her analyses of the BRFSS data and Su-Ying Liang, PhD, University of California-San Francisco, for her assistance with analyses. We are also grateful to Bonnie Davis, PhD, California Department of Health Services, for her assistance in developing and fielding the survey questions. This work was supported by a R01 grant to KAP from National Institute of Allergy and Infectious Diseases (AI43744). 346 American Journal of Preventive Medicine, Volume 24, Number 4

8 References 1. Centers for Disease Control and Prevention. Prevalence of risk behaviors for HIV infection among adults United States, MMWR Morb Mortal Wkly Rep 2001;50: Sweeney P, Fleming P, Ward J, Karon J. A minimum estimate of HIV infected persons who have been tested in the US. Paper presented at the Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Bozzette S, Berry S, Duan N, et al. The care of HIV-infected adults in the United States. HIV costs and services utilization study consortium. N Engl J Med 1998;339: Janssen R, Holtgrave D, Valdiserri R, Shepherd M, Gayle H, De Cock K. The serostatus approach to fighting the HIV epidemic: prevention strategies for infected individuals. Am J Public Health 2001;91: Myolankis E, Paliou M, Lally M, Flanigan T, Rich J. Laboratory testing for infection with the human immunodeficiency virus: established and novel approaches. Am J Med 2000;109: Respess R, Rayfield M, Dondero T. Laboratory testing and rapid HIV assays: applications for HIV surveillance in hard-to-reach populations. AIDS 2001;15(suppl 3): Kassler W. Advances in HIV testing technology and their potential impact on prevention. AIDS Educ Prev 1997;9(suppl B): Spielberg F, Kassler W. Rapid testing for HIV antibody: a technology whose time has come. Ann Intern Med 1996;125: Summers T, Spielberg F, Collins C, Coates T. Voluntary counseling, test and referral for HIV: new technologies, research findings create dynamic opportunities. J Acquir Immune Defic Syndr 2000;25(suppl 2): Spielberg F, Critchlow C, Vittinghoff E, et al. Slow diffusion of home HIV-speical collection: Provider concerns at odds with client preferences. The HIV Early Detection Study Group. Sex Transm Dis 2001;28: Spielberg F, Kurth A, Gorbach P, Goldbaum G. Moving from apprehension to action: HIV counseling and testing preferences in three at-risk populations. AIDS Educ Prev 2001;13: Spielberg F, Critchlow C, Vittinghoff E, et al. Home collection for frequent HIV testing: acceptability of oral fluids, dried blood spots and telephone results. HIV Early Detection Study Group. AIDS 2000;14: Phillips KA, Flatt SJ, Morrison KR, Coates TJ. Potential use of home HIV testing. N Engl J Med 1995;332: Branson B. Home sample collection tests for HIV infection. JAMA 1998; 280: Phillips KA, Morin S, Coates T, Fernyak S, Marsh A, Ramos-Irizarry L. Home sample collection for HIV testing. JAMA 2000;283:198 9 (letter). 16. Centers for Disease Control and Prevention. HIV counseling and testing in publicly funded sites: 1996 annual report. Atlanta, GA: U.S. Department of Health and Human Services, May Orasure Technologies, Inc. Products, infectious disease, OraSure HIV-1 Available at: Accessed on January 13, Merson MH, Feldman EA, Bayer R, Stryker J. Rapid self testing for HIV infection. Lancet 1997;349: AIDS Alert. CDC launching new era in testing by giving OK to rapid screening tests. American Health Consultants, March 1, Skolnik H, Phillips K, Binson D, Dilley J. Deciding where and how to be tested for HIV: What matters most? J Acquir Immune Defic Syndr 2001; 27: Phillips KA, Maddala T, Johnson FR. The use of conjoint analysis to measure preferences for HIV testing. Health Ser Res 2002;37: U.S. Bureau of the Census. Census 2000 data for the state of California. Available at: Accessed on March 12, Centers for Disease Control and Prevention National Health Interview Survey, U.S. Department of Health and Human Services. Available at: Accessed on May 6, Hosmer D, Lemeshow S. Applied logistic regression. New York: John Wiley & Sons, Centers for Disease Control and Prevention. HIV testing among racial/ ethnic minorities, United States, MMWR Morb Mortal Wkly Rep 2001;50: Anand G. Why rapid HIV tests, widely sold overseas, have eluded the U.S. Wall Street Journal, 2001: Phillips KA, Morrison KR, Sonnad SS, Bleecker T. HIV counseling and testing of pregnant women and women of childbearing age by primary care providers: self-reported beliefs and practices. J Acquir Immune Defic Syndr Hum Retrovirol 1997;14: Phillips KA, Maddala T, Johnson FR. Measuring preferences for health care interventions using conjoint analysis: an application to HIV testing. Health Ser Res 2002;37: Appendix Survey Questions Added to California BRFSS 1. THINK WILL BE TESTED FOR HIV IN 12 MONTHS Do you think you will be tested for HIV, the virus that causes AIDS, in the next 12 months? 1. Definitely yes 2. Probably yes 3. Maybe 4. Probably not 5. Definitely not (go to closing) 6. Don t know (go to closing) 7. Not applicable (HIV ) (go to closing) 8. Refused (go to closing) 2. PREFER TO BE TESTED FOR HIV WHERE If you were going to be tested for HIV in the next 12 months, would you prefer to get tested At a public clinic, 2. at a doctor s office, or 3. at home using a home test kit 4. No opinion (do not read) 5. None of these places (do not read) 6. Do not plan to be tested (do not read) (go to closing) 7. Don t know 8. Not applicable (HIV ) (go to closing) 9. Refused (go to closing) 3. PREFER TO BE TESTED WHERE ON BODY If you were going to be tested for HIV in the next 12 months would you prefer to get tested by Pricking your finger 2. putting a cotton swab in your mouth, or 3. urinating in a cup 4. No opinion (do not read) 5. None of these ways (do not read) 6. Do not plan to be tested (do not read) (go to closing) 7. Don t know 8. Not applicable (HIV ) (go to closing) 9. Refused (go to closing) 4. PREFER TO KNOW HIV TEST RESULTS WHEN If you were going to be tested for HIV in the next 12 months, how soon would you want your results? 1. Within 2 weeks or 2. at once, even if this means you may have to be tested again to be sure the results were correct 3. No opinion (do not read) 4. Do not plan to be tested (do not read) (go to closing) 5. Don t know 6. Not applicable (HIV ) (go to closing) 7. Refused Am J Prev Med 2003;24(4) 347

9 5. WOULD USE HOME TEST KIT HIV home testing kits may soon be available in drug stores or by mail. This kit would let you know at once if you were infected with HIV. Do you think you would use a home test kit if you were tested in the next 12 months? 1. Definitely yes 2. Probably yes 3. Maybe 4. Probably not (go to closing) 5. Definitely not (go to closing) 6. Do not plan to be tested in next 12 months (do not read) (go to closing) 7. Don t know (go to closing) 8. Not applicable (HIV ) (go to closing) 9. Refused (go to closing) 6. WOULD SPEND $50 ON HOME TEST KIT Would you buy a home test kit if it cost $50? 1. Definitely yes 2. Probably yes 3. Maybe 4. Probably not 5. Definitely not 6. Do not plan to be tested 7. Don t know 8. Not applicable (HIV ) 9. Refused 348 American Journal of Preventive Medicine, Volume 24, Number 4